Open Access, Open Data, and Open Educational Resources

Category Archives: Book Reviews

Book Review: MOOCs


Jonathan Haber, MOOCS. The MIT Press Essential Knowledge Series. Cambridge, Mass. : The MIT Press, 2014.

I read Jonathan Haber’s book MOOCs a few months ago, and am glad to finally offer some thoughts. Despite a remarkable cooling of interest in MOOCs, there are still plenty of reasons to consider what role they might play in higher education. Haber, perhaps best known for his year-long MOOC experiment to obtain the equivalent of a bachelor’s degree, here offers a readable and balanced account of the MOOC environment.

Haber begins by outlining the history of MOOCs (massive open online courses), pointing out that “open” was an earlier driver than “massive” with MIT’s OpenCourseWare initiative for class materials (begun in 2002), though many of those courses lack video lectures. The first real MOOC came along in 2008, “Connectivism and Connective Knowledge,” taught by Stephen Downes and George Siemens. In the connectivist model, class size became an asset, not a liability (p. 39):

For the bigger the connectivist “class,” the greater the potential for the quantity and variety of nodal connections that define success for networked learning.

However, as MOOCs evolved, most were not designed around a specific pedagogical method, and Haber notes how different the learning experience is between connectivist and non-connectivist MOOCs. A tool for student connection common to both models is the discussion board, though they can be overwhelming to students, resulting in low participation rates. Scheduled vs. on-demand MOOCs have different types of discussion, with the latter focusing more on test and assignment support rather than on general course topics. Haber provides an interesting analysis of other ways that scheduled and on-demand MOOCs differ (p. 78-79).

In his chapter Issues and Controversies (p. 89-131), Haber first focuses on the low completion rates of MOOCs (a problem shared by a MOOC I wrote about last year). He argues that MOOC sign-ups are due mostly to curiosity rather than commitment. Still, though completion rates may be low, the raw completion numbers are still very large, and Haber quotes a professor who remarks that the number of students completing his MOOC is equal to all of the students he has taught in his career up to that point. Problems such as course demand level, cheating, plagiarism, and student identity are being addressed in a variety of ways, such as Coursera’s signature track identity verification.

On the positive side, there’s evidence that the shorter lectures used in most MOOCs are more effective, and that the ability to change speed, pause, and repeat lectures has a pedagogical impact. The interaction of older and younger learners common in MOOCs is rare in traditional education. The modularity of MOOCs is increasingly being utilized, and MOOCs have been successful in blended learning, rather than as a substitute for the classroom. Indeed, edX material is used at MIT to flip courses, and there’s extensive discussion about how MOOCs can fit into the flipped classroom model (p. 156-161). On the whole, MOOCs have raised the bar for online education in terms of production value, creativity, and risk-taking.

In these days of corporate open-washing, anything claiming to be open bears further examination. Haber notes that “open” tends to be interpreted by the public as “free,” despite the need in some MOOCs to purchase materials in order for the student to benefit the most from the course. Haber offers solid discussions of intellectual property (beginning on p. 118) and openness (beginning p. 122). A central problem has been that academic libraries license content for their campuses which cannot legally be shared with large numbers of unaffiliated students. Additionally, educational use is not automatically fair use (a common misunderstanding). Options for using external material include a full fair use analysis, obtaining permission (often at a cost), linking to content, and/or using openly licensed resources. And of course, most MOOCs are not openly licensed themselves. However, edX seems to be upholding open values and thriving, according to a recent article.

Haber also covers the difficulties involved in getting credit for MOOC courses from institutions of higher learning through programs like high school Advanced Placement (AP), the College Board’s College Level Examination Program (CLEP), and the American Council of Education’s (ACE) CREDIT program, which accredits courses for college-level equivalency. Publicity and incentives for the one-off alternative credit are not sufficient, which may explain why there were no sign-ups for either an ACE transcript for a MOOC or a Udacity-Colorado State course in computer programming (p. 106). Yet the future of MOOCs for younger learners, Haber says, may be alongside these existing programs.

This book introduced me to Straighterline and the SPOC (small private online course- for example, CopyrightX, which I hope to take), but the MOOC environment is apparently so fast-moving that some interesting initiatives are now defunct, such as MOOCs Forum, MOOC Campus and Haber perhaps overstates the altruistic purposes of MOOCs (p. 187), and his statements about the cost challenge of MOOCs to residential education may be premature.

MOOCs is part of the MIT Press Essential Knowledge series, which notably includes Peter Suber’s Open Access and John Palfrey’s Intellectual Property Strategy (which I reviewed previously). In addition to an index and notes, it includes a glossary, additional resources, and a list of MOOC providers. It’s an enjoyable and informative read, though not one inspiring certainty, perhaps best communicated by one last heavily-qualified quote (p. 194):

But if MOOCs continue to embrace-or even expand on- the culture of experimentation and innovation that has already set them apart from nearly all other adventures in technology-based learning, if they continue to offer high-quality free teaching to the world while also serving as the laboratory where educational innovation thrives, then whatever MOOCs are today or whatever they evolve into, they are likely to leave an important mark on whatever ends up being called higher education in the future.

Book Review: Issues in Open Research Data

Issues in Open Research Data

Moore, Samuel A. (ed.), Issues in Open Research Data (London: Ubiquity Press, 2014).

Bringing together contributed chapters on a wide variety of topics, Issues in Open Research Data is a highly informative volume of great current interest. It’s also an open access book, available to read or download online and released under a CC BY license. Three of the nine chapters have been previously published, but benefit from inclusion here. In the interest of full disclosure, I’m listed as a book supporter (through in the initial pages.

In his Editor’s Introduction, Samuel A. Moore introduces the Panton Principles for data sharing, inspired by the idea that “sharing data is simply better for science.” Moore believes each principle builds on the previous one:

  1. When publishing data, make an explicit and robust statement of your wishes.
  2. Use a recognized waiver or license that is appropriate for data.
  3. If you want your data to be effectively used and added to by others, it should be open as defined by the Open Knowledge/Data Definition— in particular, non-commercial and other restrictive clauses should not be used.
  4. Explicit dedication of data underlying published science into the public domain via PDDL or CC0 is strongly recommended and ensures compliance with both the Science Commons Protocol for Implementing Open Access Data and the Open Knowledge/Data Definition.

In “Open Content Mining” Peter Murray-Rust, Jennifer C. Molloy and Diane Cabell make a number of important points regarding text and data mining (TDM). Both publisher restrictions and law (recently liberalized in the UK) can block TDM. And publisher contracts with libraries, often made under non-disclosure agreements, can override copyright and database rights. This chapter also includes a useful table of the TDM restrictions of major journal publishers. (Those interested in exploring further may want to check out ContentMine.)

“Data sharing in a humanitarian organization: the experience of Médecins Sans Frontières” by Unni Karunakara covers the development of MSF’s data sharing policy, adopted in 2012 (its research repository was established in 2008). MSF’s overriding imperative was to ensure that patients were not harmed due to political or ethnic strife.

Sarah Callaghan makes a number of interesting points in her chapter “Open Data in the Earth and Climate Sciences.” Because much of earth science data is observational, it is not reproducible. “Climategate,” the exposure of researcher emails in 2009, has helped drive the field toward openness. However, there remain several barriers. The highly competitive research environment causes researchers to hoard data, though funder policies on open data are changing this. Where data has commercial value, non-disclosure agreements can come into play. Callaghan notes the paradox that putting restrictions on collaborative spaces makes sharing more likely (the Open Science Framework is a good example). She also shares a case in which an article based on open data was published three years before the researchers who produced the data published. It is becoming likely that funders will increasingly monitor data use and require acknowledgement of data sources if used in a publication. Data papers (short articles describing a dataset and the details of collection, processing, and software) may encourage open data. Researchers are more likely to deposit data if given credit through a data journal. However, data journals need to certify data hosts and provide guidance on how to peer review a dataset.

In “Open Minded Psychology” Wouter van den Bos, Mirjam A. Jenny, and Dirk U. Wulff share a discouraging statistic: 73% of corresponding authors failed to share data from published papers on request. A significant barrier is that providing data means substantial work. Usability can be enhanced by avoiding proprietary software and following standards for structuring data sets (an example of the latter is OpenfMRI). The authors discuss privacy issues as well, which in the case of fMRI includes a 3D image of the participant’s face. The value of open data is that data sets can be combined, used to address new questions, analyzed with novel statistical methods, or used as an independent replication data set. The authors conclude:

Open science is simply more efficient science; it will speed up discovery and our understanding of the world.

Ross Mounce’s chapter “Open Data and Palaeontology” is interesting for its examination of specific data portals such as the Paleobiology Database, focusing in particular on the licensing of each. He advocates open licenses such as the CC0 license, and argues against author choice in licensing, pointing out that it creates complexity and results in data sharing compatibility problems. And even though articles with data are cited more often, Mounce points out that traditionally indexing occurs only for the main paper, not supplementary files where data usually resides.

Probably the most thought-provoking yet least data-focused chapter is “The Need to Humanize Open Science” by Eric Kansa of Open Context, an open data publishing venue for archaeology and related fields. Starting with open data but mostly about the interaction of neoliberal policies and openness, the chapter deserves a more extensive analysis than I can give here, but those interested in the context against which openness struggles may want to read his blog post on the subject, in addition to this chapter.

Other chapters cover the role of open data in health care, drug discovery, and economics. Common themes include:

  • encouraging the adoption of open data practices and the need for incentives
  • the importance of licensing data as openly as possible
  • the challenges of anonymization of personal data
  • an emphasis on the usability of open data

As someone without a strong background in data (open or not), I learned a great deal from this book, and highly recommend it as an introduction to a range of open data issues.

Book Review: Open Access and the Humanities

Open Access and the Humanities

Martin Paul Eve, Open Access and the Humanities: Contexts, Controversies and the Future (Cambridge: Cambridge University Press, 2014).

Martin Eve’s new book is a welcome examination of the unique challenges that the humanities face in open access publishing. Appropriately enough, it is published open access by Cambridge University Press (as separate PDFs, or download the full PDF from the Internet Archive) under a Creative Commons Attribution-ShareAlike license (CC BY-SA 4.0). Dr. Eve is a Lecturer in English at the University of Lincoln, co-directs the Open Library of the Humanities (OLH), and frequently speaks and blogs about open access.

Some of the challenges for the humanities in open access publishing become apparent in Peter Suber’s preface. In addition to a general lack of funding in these fields, humanities journals tend to have high rejection rates, so article processing charges (APCs) become a non-starter.  The fee-based model works best in well-funded fields with relatively low rejection rates.  In addition, it is books, not journal articles, that are of greatest importance.

Therefore it is no surprise that Eve’s chapters on economics and monographs (Chapters 2 and 4) serve as the heart of the book. Eve hypothesizes three economic models (p. 57) in which authors are paid for their work, publishers are paid for their services, and libraries provide cooperative funding. Those familiar with the OLH’s funding model won’t be surprised to find the first two models dismissed in short order, and the cooperative model examined at length. Indeed, the cooperative model has already been used to support arXiv, SCOAP3, and most recently Knowledge Unlatched, albeit in a pilot effort. In considering this model, Eve begins by stating that there is already enough money in academic publishing to cover the current production of articles and monographs. Problems of transition funding are addressed, though to me the most interesting problem is how to handle free riders in cooperative efforts (p.75):

The first of these difficulties, the so-called ‘free-rider’ problem, relates to the economic understanding that rationally self-interested actors do not wish to pay for commodities from which others benefit for free. In other words, except in philanthropic modes or systems of taxation for public good, most people usually resist paying for goods for which only they pay, but from which non-purchasers also derive benefit. This results, for gold open access publishing, in a kind of prisoner’s dilemma. If all library entities behave in a purely self-interested way and disallow free riders, these collectively underwritten, non-APC models cannot emerge. Admittedly, the increasing enclosure of universities within market logics doubtless makes it harder for acquisition librarians to justify expenditure on projects where there are free riders to senior managers.

I think there may be additional problems for libraries here (enough for a separate blog post), and I hoped for a fuller examination of this voluntary contribution model, though recent efforts, as Eve notes, have been successful.

Chapter 3 on open licensing helpfully outlines the three current types of copyright and licensing agreements authors will encounter: full copyright transfer, an exclusive license to publish, and a non-exclusive license to publish (p. 88). This chapter also does a good job of describing how copyright restricts the uses of research, and discusses some common objections to CC licenses, such as concerns about derivatives, undesirable re-use, and commercial use.

Eve explains the economic problems of monograph publishing in chapter 4 and offers an overview of four current models of providing open access to them (p. 130-135): print subsidy, institutional subsidy, freemium, and collective funding. I was unaware that publishers pay for peer review of monographs, which contributes to their higher production costs and resulting book processing charges (BPCs) that are largely unaffordable for humanities authors. I was glad to see two issues addressed in this chapter. The first is the role of commercial presses in determining what is published (and therefore who is hired or receives tenure) based on evaluations of potential sales. The second is the role of scholarly societies (p. 128):

…calls to protect society revenue models are often inextricable from calls to protect publisher profits; the two are interwoven. This rhetoric of economy and sustainability, it must also not be forgotten, will always make one group’s sustainability possible only at the expense of another: usually the library.

There are interesting discussions of the advantages and disadvantages of archiving (“green” open access) in the humanities, though it seems that there are more of the latter. Archiving is often not permitted for monographs, pagination is lost, and the strain on library budgets goes unaddressed. Strangely, Eve omits the biggest problem, which is that a majority of authors have permission to archive journal articles, yet so few actually take advantage of it. Additionally, in proposing an archiving workflow for authors, Eve suggests checking SHERPA/RoMEO for permissions. While that is an invaluable resource, I think it is more important that authors read (or at least quickly review) their publishing contract for archiving rights.

Eve’s final chapter concerns innovations, and mostly addresses peer review. While emphasizing that open access does not require changes in peer review, he argues that the opportunity to reconsider other publishing practices should not be missed, and in particular to address the false perception that open access means lower quality. Critiques of “filter-first” peer review are echoed. Two explorations in this chapter are especially interesting: first, how the PLOS One type of review for technical correctness rather than importance might be translated to humanities fields; and second, how assumptions that speed and precedence are necessary only for the sciences might be mistaken.

Eve’s writing style is clear, though over-use of comma phrases tended to bog this reader down at times, and there is the occasional odd construction (“greenly depositing,” p. 11). The book ends somewhat abruptly and might have benefited from a concluding chapter to help synthesize the many aspects of open access publishing for the humanities. The glossary (p. 179-181), while helpful in itself, defines some terms so minimally as to lead to further questions. In addition to the glossary and Suber’s preface, the book contains notes, a bibliography, and an index (which can be supplemented by searching the PDF version).

Despite these minor criticisms, Open Access and the Humanities is thought-provoking and remarkably balanced, perhaps due to Eve’s dual role as open access advocate and publisher. Eve approaches all of these complex issues in a spirit of philosophical investigation, and does not avoid examination of related issues such as academic freedom and research assessment. A broad audience of humanists, publishers, and librarians will find value in this exploration of open access for humanities disciplines, and many of them will no doubt be cheering for the success of the new publishing models it describes, such as the Open Library of the Humanities.

Book Review: The Virtues of Openness

The Virtues of Openness

The Virtues of Openness: Education, Science, and Scholarship in the Digital Age by Michael A. Peters and Peter Roberts (2012) will appeal to anyone interested in open movements in respect to academia. The book includes eight chapters (three of which were previously published), an introduction, postscript, and extensive references. The authors are both professors of education in New Zealand (Peters at the University of Waikato; Roberts at the University of Canterbury).

In addition to exploring the many aspects of openness (which makes defining it so difficult), the authors make an important point that bears remembering when we are tempted by binary conceptions such as open or not (Introduction, p.6):

All open systems have limits, and there are limits to openness– limits to “open” markets, to open societies, to open code.

It’s a theme the authors return to repeatedly, particularly in the context of the philosophy of education, also noting that these limits can serve positive functions.

Chapter 4, “Open Education and Open Knowledge Production” (p. 55-76) covers the serials crisis and open access with the greatest depth, but I learned the most from Chapter 2, “The Philosophy of Open Science” (p. 30-42), and Chapter 3, “Openness as an Educational Virtue” (p. 43-54). Chapter 2 begins by emphasizing the narratives of openness in the West and their relation to Enlightenment thought, in particular the ways in which openness is freedom. The bulk of the chapter goes on to consider philosophies of openness from various thinkers. Of particular interest are the connections between thinkers leading up to the concept of open access for scholarly literature. Karl Popper (author of The Open Society And Its Enemies) was a strong influence on George Soros, whose Open Society Institute (now the Open Society Foundations) was the driving force behind the Budapest Open Access Initiative. Here the political conception of an open society contains both the market and science as primary institutions based on shared values of freedom and truth, though it’s worth noting how often these institutions are in conflict today. Indeed, the authors make this clear in their summary of Chapter 4 (p. 76):

In essence, the open knowledge economy provides a completely different model to the neoliberal knowledge economy and also challenges the underlying neoliberal ideas of ownership, authorship, human capital, and intellectual property rights as well as principles of the access, distribution, and creation of knowledge.

In chapter 3 (“Openness as an Educational Virtue”), a philosophy of openness in pedagogy focuses on the work of Brazilian educationist Paulo Freire. Openness, the authors contend, includes but is not limited to open-mindedness, and is contrasted with forms of closure such as “dogmatism, excessive certainty, and an unreflective rejection of either the old or the new” (p. 44). Throughout his career, Freire identified human characteristics of value in teaching and learning situations, such as “humility, the ability to listen, showing care and respect for those with whom we work in educational settings, tolerance, an inquiring and investigative frame of mind, and a willingness to take risks” (p. 49). For Freire, openness is a permanent orientation to life itself, recognizing that we are always unfinished beings.

The authors view the university emphasis on performance as a form of closure in Chapter 5 (“Scholarly Publishing and the Politics of Openness: Knowledge Production in Contemporary Universities”), stating that “it is performance, not knowledge, that counts” (p. 82, 83):

Performance, as measured by lists of “outputs,” becomes the accepted substitute for knowledge and is seen as translatable across individuals, departmental groupings, disciplines, and institutions.

And these outputs must be quantified (p. 84):

…research activity counts only insofar as it is measurable. Behind this trend lies a quest for certainty, a discomfort with that which is complex or messy, and an inability to deal with the immeasurable.

Although the ongoing revolution in scholarly communication also relies in part on measurables such as review and altmetric scores, it shows a willingness to deal with uncertainty and the immeasurable through open peer review and post-publication peer review. The authors identify an interesting casualty of the culture of performativity, which is time, or the lack of it (p. 87). One problem not addressed here is that being open is more time-consuming in the current publishing environment. Those who want to be open must make additional effort, whether it is seeking out an open access journal, archiving their manuscript, or organizing, describing, and providing access to their data. Clearly norms must change so that closure is not the easy, time-saving alternative. Time pressures also drive quantification in academic evaluation, since it is faster to look at scores than to read someone’s scholarship.

At times The Virtues of Openness shows considerable overlap in chapter topics; at other times the transitions between chapters are jarring, perhaps because some were previously published. As such, the book feels like a collection of chapters rather than a connected narrative. Also, a few of the figures and tables are either not particularly enlightening (“Applications of openness” on p. 67) or outdated (2001 scientific publishing market players, p. 58).

However, this volume is a solid resource for those interested in exploring the thinkers who have contributed to the philosophy of openness in a variety of disciplines. The authors do an admirable, mostly jargon-free job of introducing and clarifying different aspects of openness, and of emphasizing its limits.

The Virtues of Openness is available in Newman Library and from the publisher.

Book Review: Reclaiming Fair Use

Reclaiming Fair Use Reclaiming Fair Use: How to Put Balance Back in Copyright by Patricia Aufderheide and Peter Jaszi was published by the University of Chicago Press in 2011. It’s a well-written history of fair use interpretation and an important corrective to over-cautiousness in asserting user rights. Fair use is a provision of U.S. copyright law that, broadly speaking, allows use of copyrighted works when the social benefit is greater than the owner’s loss. The law sets out four factors which are used to determine whether fair use can be employed: the nature of the use, the nature of the work used, the extent of the use, and its potential economic effect. But since there is no bright line or definitive calculation of the four factors (and other factors which may have bearing), the effect has been limiting (p. xi):

We saw that when people do not understand the law, when they are constantly afraid that they might get caught for referring to copyrighted culture- whether an image, or a phrase of a song, or a popular cartoon character- they can’t do their best work.

Aufderheide and Jaszi feel that the four factors (and checklists based on them) have been hindrance (p. 183):

People love checklists, because they hope that the lists will do their fair-use reasoning for them. But checklists tend to be more trouble than help. Sometimes a checklist simply discourages fair use in situations where the user might have an adequate rationale not captured by the list. More often, checklists simply lead to further confusion. Focused on the four factors, they treat the factors as if they had a concreteness that they do not. Those four factors have been widely interpreted by judges over the years.

Instead they distill fair use evaluation into three questions (p. 24 and 135):

Was the use of copyrighted material for a different purpose, rather than just reuse for the original purpose? Was the amount of material taken appropriate to the purpose of the use? Was it reasonable within the field or discipline it was made in?

The first and third questions are especially important in the revitalization of fair use. While copyright has become “long and strong” in recent decades, fair use has made a comeback since the late 1990s to lend the law more balance. Fair use interpretations have been primarily strengthened in two ways: first through the concept of transformativeness (use for a different purpose than originally intended), and more recently through development of codes of practice for particular fields. Both are now major considerations by courts (p. 80). Aufderheide and Jaszi have been leaders in developing best practices for various communities, first with documentary filmmakers (a process related in Chapter 7) and most recently as contributors to initial work toward a code of practice for the visual arts (PDF).

Codes of best practice “represent a common understanding in a community of practice” (p. 120) and emphasize demonstrating good faith (e.g. through attribution). The codes developed thus far are in agreement on three areas of fair use: critique, illustration, and incidental capture. The codes are also balanced in the sense that the communities (e.g. documentary filmmakers) are often creators as well, so they must take into account how their own work might be used. Aufderheide and Jaszi emphasize that, like a muscle, fair use is strengthened by use– it is one arena in which behavior affects the law, not vice-versa. In addition to communities of practice, the law provides exceptions for certain kinds of use, such as the educational exemptions in Section 110-1 and 110-2.

While the authors champion fair use, they are clear about the problems that remain. In the digital environment, many works are leased rather than owned, and contracts may include language limiting fair use rights. The Digital Millennium Copyright Act (DMCA) of 1998 made it illegal to override digital encryption, so exercising one’s fair use rights becomes impossible. Reliance on the courts to interpret fair use has its disadvantages, and one casualty has been music sampling. The interaction of three court cases has severely limited fair use for music (p. 90-93). Formal copyright registration entitles owners to statutory damages, and the potential maximum has a chilling effect (p. 32). The courts have also expanded secondary liability. The authors call for for advocacy on DMCA reform as well as on orphan works.

Aufderheide and Jaszi are unexpectedly critical of free-culture and commons advocates. They indict free-culture activists for making copyright the villain (p. 48) and seeking alternatives elsewhere rather than acknowledging balancing effects of copyright law such as fair use (p. 54):

The commons rhetoric… celebrates a particular vision of the public domain as a space entirely free of intellectual property constraint, while either ignoring or slighting exemptions and balancing features that limit copyright owners’ monopoly control.

Yet the commons is growing steadily, and search engines now allow users to filter images by license. And in their discussion of the public domain (p. 141), the authors fail to mention the Creative Commons Zero (CC0) license for intentionally placing works in the public domain. While commons advocates may have overlooked fair use, the unnecessary distinction between the two approaches is contradicted by the authors’ own work on a code of practice for OpenCourseWare, which relies on both open licensing and fair use.

The international environment for fair use is covered in Chapter 10. Most countries lack a fair use provision, but have a much lower risk of litigation and lack statutory damages for infringement. Because fair use is the exception rather than the rule, harmonization of copyright through treaties is a continuing threat to it.

Fair use is deliberately vague, and always a case-by-case decision. To Aufderheide and Jaszi, this is a feature, not a bug (p. 163):

Creators benefit from the fact that the copyright law does not exactly specify how to apply fair use…. Fair use is flexible; it is not uncertain or unreliable.

Reclaiming Fair Use features inset boxes throughout the text, “Fair Use: You Be The Judge” (with answers at the back) and “True Tales of Fair Use,” and has five useful appendices, including a template for a code of best practices and a section on myths and realities of fair use. While it contains more background than some readers may desire (they can go straight to Chapter 9, “How To Fair Use”), this book is a valuable perspective on fair use and always interesting and well-written.

More information about fair use, including codes of practice, can be found at the Center for Media & Social Impact at American University, which Aufderheide co-directs. In addition, Jaszi provided testimony on fair use to a House of Representatives subcommittee in January (his testimony begins at 39:00 in the video, and his written submission is available in PDF).

Reclaiming Fair Use is available as an e-book through the University Libraries.

Open@VT on Mastodon

Loading Mastodon feed...